Analytic Methods for Constructing Cross-Sectional Profiles of Health Care Providers

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

National effort is currently directed toward developing and disseminating comparative information involving both outcomes and processes of care for health care providers. Univariate provider-specific comparative indices based on posterior summaries as well as indices based on maximum likelihood estimates have been developed for use in the cross-sectional setting. A remaining issue in the dissemination of cross-sectional profiles relates to the multivariate nature of the indices: often many performance measures are used to assess quality for a particular provider. Because this information can often be contradictory and overwhelming, there is a need for measures that summarize quality at a provider level. This article proposes the use of latent variable models for comparing health care providers in the cross-sectional setting when each provider is measured on more than one dimension of care. By combining information across dimensions of care within a provider, an integrated analysis can produce a composite measure of quality and has more statistical power to detect differences among providers. As the number of individual performance measures grows over time, composite measures will become increasingly important tools to support decision making by consumers, payors, and providers.